Model comparison
GPT-5.5 Instant vs Step 3.7 Flash
GPT-5.5 Instant is the stronger model overall, scoring 42.7 to 37.3 on the Noometry Index.
Last verified . 2 shared benchmarks.
StepFunSummary
- They share 2 benchmarks with published results for both. GPT-5.5 Instant scores higher in 2 categories and Step 3.7 Flash in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3.7 Flash leads 42.9 to 26.5.
- The biggest single-benchmark swing is SciCode: 48.6% for GPT-5.5 Instant and 40% for Step 3.7 Flash.
- Step 3.7 Flash has downloadable open weights; the other is API-only.
Side by side
| GPT-5.5 Instant | Step 3.7 Flash | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 42.7 | 37.3 |
| Released | 2026-05-05 | 2026-05-29 |
| Weights | Proprietary | Open |
| Context window | — | 256K |
| Max output | — | 256K |
| Input $ / M tokens | — | $0.18 |
| Output $ / M tokens | — | $1.11 |
| Results tracked | 27 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 Instant leads
GPT-5.5 Instant: 44.3 (#74), Step 3.7 Flash: 40.0 (#150)
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| SciCode | 48.6% | 40% |
| LMArena Coding | 1433 | — |
| ALE-Bench | — | 694.12 |
Reasoning GPT-5.5 Instant leads
GPT-5.5 Instant: 24.9 (#155), Step 3.7 Flash: 21.6 (#219)
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| CritPt | 0% | 2.3% |
| NYT Connections (extended) | — | 39.7% |
| Chess Puzzles | 12% | — |
| LMArena Hard Prompts | 1426 | — |
| Epoch Capabilities Index | 142.52 | — |
Math Step 3.7 Flash leads
GPT-5.5 Instant: 26.5 (#259), Step 3.7 Flash: 42.9 (#82)
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 26.3% | — |
| FrontierMath Tier 4 | 2.4% | — |
| MathArena Final-Answer Competitions | — | 68.5% |
| OTIS Mock AIME 2024-2025 | 68.1% | — |
| LMArena Math | 1420 | — |
Knowledge Not comparable
GPT-5.5 Instant: 48.9 (#74), Step 3.7 Flash: —
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| GPQA Diamond | 82.5% | — |
| LMArena Expert | 1409 | — |
Multimodal Not comparable
GPT-5.5 Instant: 40.0 (#52), Step 3.7 Flash: —
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| LMArena Vision | 1250 | — |
| LMArena Document | 1403 | — |
Multilingual Not comparable
GPT-5.5 Instant: 52.8 (#80), Step 3.7 Flash: —
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1417 | — |
| LMArena Chinese | 1456 | — |
| LMArena French | 1428 | — |
| LMArena German | 1411 | — |
| LMArena Japanese | 1408 | — |
| LMArena Korean | 1392 | — |
| LMArena Russian | 1431 | — |
| LMArena Spanish | 1429 | — |
Instruction Following Not comparable
GPT-5.5 Instant: 74.2 (#100), Step 3.7 Flash: —
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1406 | — |
Long Context Not comparable
GPT-5.5 Instant: 43.4 (#96), Step 3.7 Flash: —
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| LMArena Longer Query | 1422 | — |
Writing & Preference Not comparable
GPT-5.5 Instant: 61.8 (#85), Step 3.7 Flash: —
| Benchmark | GPT-5.5 Instant | Step 3.7 Flash |
|---|---|---|
| LMArena Text | 1419 | — |
| LMArena Creative Writing | 1419 | — |
| LMArena Multi-Turn | 1433 | — |
Frequently asked questions
Is GPT-5.5 Instant better than Step 3.7 Flash?
GPT-5.5 Instant is the stronger model overall, scoring 42.7 to 37.3 on the Noometry Index.
Is GPT-5.5 Instant or Step 3.7 Flash better for coding?
GPT-5.5 Instant scores higher on coding benchmarks: 44.3 versus 40.0 in the Noometry coding category.
How many benchmarks do GPT-5.5 Instant and Step 3.7 Flash share?
2 benchmarks have published results for both models. GPT-5.5 Instant has 27 scored results on Noometry and Step 3.7 Flash has 5.